System and method for graph encoding of physical activity data

Abstract

A method for determining properties associated with a graph is disclosed herein. The graph includes a plurality of nodes and a plurality of connections between the nodes. The method comprises receiving global positioning system (GPS) data points for a plurality of GPS tracks associated with the graph. The method further comprises mapping the GPS data points for each of the plurality of the GPS tracks to the graph by determining an activity track based on the GPS track, the activity track including a plurality of nodes from the graph and at least one connection between the nodes. Additionally, the method comprises creating a path matrix including a plurality of values, each value in the path matrix associated with (i) one of the plurality of connections between nodes of the graph and (ii) one of the determined activity tracks.

What is claimed is: 
     
       1. A method for identifying an edge of a graph associated with selected demographic data from a user database, the method comprising:
 defining the graph within a bounded polygon, the graph including a plurality of nodes and a plurality of connections between the nodes; 
 receiving global positioning system (GPS) data points for a plurality of GPS tracks from a plurality of GPS-enabled devices; 
 mapping the GPS data points for each of the plurality of the GPS tracks to the graph by determining an activity track based on the GPS track, the activity track including a plurality of nodes from the graph and at least one connection between the nodes; 
 creating a path matrix including a plurality of values, each value in the path matrix associated with (i) one of the plurality of connections between nodes of the graph and (ii) one of the determined activity tracks; 
 defining a user profile vector based at least in part on the selected demographic data in the user database; 
 multiplying the path matrix and the user profile vector to obtain a resulting vector; and 
 determining at least one edge of the graph associated with selected demographic based on the resulting vector. 
 
     
     
       2. The method of  claim 1  wherein the resulting vector includes properties indicative of frequency of traversal of each of the plurality of connections of the graph in association with the selected demographic data. 
     
     
       3. The method of  claim 2  wherein each of the activity tracks is associated with user profile data in the user database, and wherein the user profile vector is defined based on the user profile data. 
     
     
       4. The method of  claim 3  wherein the user profile data includes user gender and type of activity. 
     
     
       5. The method of  claim 1  wherein defining the graph within the bounded polygon includes forming a bounding box based at least in part on a maximum and a minimum of a longitude and a latitude of the GPS data points for at least one GPS track. 
     
     
       6. The method of  claim 1  wherein defining the graph within the bounded polygon includes determining a simplified GPS sequence and forming the bounded polygon based on the simplified GPS sequence. 
     
     
       7. The method of  claim 1  wherein receiving GPS data points for the plurality of GPS tracks occurs prior to defining the graph within the bounded polygon. 
     
     
       8. A non-transitory computer readable medium containing instructions for identifying an edge of a graph associated with selected demographic data from a user database by:
 defining the graph within a bounded polygon, the graph including a plurality of nodes and a plurality of connections between the nodes; 
 receiving global positioning system (GPS) data points for a plurality of GPS tracks from a plurality of GPS-enabled devices; 
 mapping the GPS data points for each of the plurality of the GPS tracks to the graph by determining an activity track based on the GPS track, the activity track including a plurality of nodes from the graph and at least one connection between the nodes; 
 creating a path matrix including a plurality of values, each value in the path matrix associated with (i) one of the plurality of connections between nodes of the graph and (ii) one of the determined activity tracks; 
 defining a user profile vector based at least in part on the selected demographic data in the user database; 
 multiplying the path matrix and the user profile vector to obtain a resulting vector; and 
 
       determining at least one edge of the graph associated with selected demographic based on the resulting vector. 
     
     
       9. The computer readable medium of  claim 7  wherein the resulting vector includes properties indicative of frequency of traversal of each of the plurality of connections of the graph in association with the selected demographic data. 
     
     
       10. The computer readable medium of  claim 8  wherein each of the activity tracks is associated with user profile data in the user database, and wherein the user profile vector is defined based on the user profile data. 
     
     
       11. The computer readable medium of  claim 9  wherein the user profile data includes user gender and type of activity. 
     
     
       12. The computer readable medium of  claim 7  wherein defining the graph within the bounded polygon includes defining a bounding box based at least in part on a maximum and a minimum of a longitude and a latitude of the GPS data points for at least one GPS track. 
     
     
       13. The computer readable medium of  claim 7  wherein receiving GPS data points for the plurality of GPS tracks occurs prior to defining the graph within the bounded polygon. 
     
     
       14. A method for determining properties associated with a graph, the graph including a plurality of nodes and a plurality of connections between the nodes, the method comprising:
 receiving global positioning system (GPS) data points for a plurality of GPS tracks associated with the graph; 
 mapping the GPS data points for each of the plurality of the GPS tracks to the graph by determining an activity track based on the GPS track, the activity track including a plurality of nodes from the graph and at least one connection between the nodes; and 
 creating a path matrix including a plurality of values, each value in the path matrix associated with (i) one of the plurality of connections between nodes of the graph and (ii) one of the determined activity tracks; 
 performing a mathematical operation on the path matrix in order to determine at least one property for the graph, the mathematical operation including multiplying the path matrix by a first vector to provide a second vector, the second vector having properties indicative of frequency of traversal of the plurality of connections of the graph. 
 
     
     
       15. The method of  claim 14  wherein each of the activity tracks is associated with profile data, and wherein the first vector is derived based on the profile data. 
     
     
       16. The method of  claim 15  wherein the profile data includes user gender and type of activity. 
     
     
       17. The method of  claim 14  wherein the at least one property associated with the graph is indicative of (i) a frequency of traversal of the plurality of connections of the graph for a selected demographic of users or (ii) activity tracks that utilized a particular connection between the nodes of the graph. 
     
     
       18. The method of  claim 14  wherein the graph is bounded within a bounding box that includes the plurality of nodes.